It’s tempting to treat AI as the reason brand consistency has gotten harder to control. But it isn’t. As Ed Miller, Creative Director at CBIZ (the 8th-largest accounting and professional services firm in the US, with over 9,500 employees and 135,000 clients) put it plainly in a recent conversation on creative operations: brand inconsistencies don’t begin with AI. They already existed. AI just moves faster than the gaps in your systems can keep up with.
We’re calling it the DAM AI Governance Gap.
What is the DAM AI Governance Gap?
The DAM AI Governance Gap is the distance between how fast your organization is adopting AI and how ready your underlying digital asset management, metadata, and brand governance actually are to support it.
It’s not a technology gap; it’s a readiness gap. AI doesn’t create brand inconsistency on its own; it finds every existing crack in your governance and widens it, faster than any human ever could.
On the other hand, if you can close that gap, AI becomes an accelerant for brand-compliant work at scale.
CBIZ learned this the hard way. In the middle of a merger and a rebrand, the creative team found themselves fighting a familiar fire: teams using templates from two brands ago, unapproved logos sneaking into decks, and no single source of truth for something as basic as a partner’s current headshot. The instinct was to treat it as a process problem: tighten reviews, remind people of the rules, move faster. It took stepping back to realize the real issue was actually structural.
The Real Problem Usually isn’t Process, it’s the System
There’s an important distinction between a process problem and a system problem.
A process problem is local: a team’s workflow needs tightening, a step is missing, a handoff is unclear. A system problem is bigger: it’s about governance, ownership, and how work actually moves across an entire organization.
CBIZ’s post-merger chaos looked like a process problem on the surface. But when Miller took a step back, the pattern was unmistakable: assets scattered across multiple legacy servers inherited from a prior merger, a legacy DAM some teams used heavily and others had never logged into, and template tools that couldn’t keep up with real demand. As Miller described it, “if I was going to design a system for how we were going to run from a blank slate, this is not how I would have designed it.” That’s the gut check that separates a process fix from a system fix.
The pattern underneath it all is worth naming directly: rogue design.
Rogue design means people going around the creative team and building their own assets. It’s a sign that the compliant path is too slow, too rigid, or doesn’t support what they actually need to do. When a system doesn’t meet people’s real demands, they build workarounds.
This is the DAM AI Governance Gap in its rawest form—a governance problem that existed before anyone introduced a generative AI tool into the mix. AI doesn’t open this gap. It just travels through it a lot faster than a person copying a logo into PowerPoint ever could.
What Does Good Brand Governance Actually Look Like?
Good brand governance doesn’t require more rules; it’s removing the reasons people go looking for a workaround. Making the correct branding decision should be the easiest decision available to users, not the hardest one.
In practice, that’s meant locking down templates so formatting can’t be broken, while still giving non-designers enough flexibility to get their work done without filing a request and waiting on a designer.
Done well, this doesn’t just reduce off-brand output. It can remove the approval bottleneck entirely: if a template genuinely can’t be broken, there’s no reason to route every use of it through manual review.
Where AI Fits. And Where it Doesn’t (Yet)
This is where AI enters the picture, and where the opportunity and the exposure split cleanly into two very different use cases.
The opportunity: AI-assisted ideation inside a creative team’s own workflow should speed up the early, open-ended part of a design problem where you’re throwing ideas at the wall before refining. That’s a meaningfully different, lower-risk application than AI generating brand-facing content directly.
The exposure: general-purpose AI tools tend to treat brand guidelines as suggestions rather than rules. As Miller described it, it’s a lot like the pirate code from Pirates of the Caribbean: “more like guidelines anyway, as opposed to strict rules.” Most users don’t have the AI literacy to consistently prompt around that, and non-designers in particular often can’t tell when the output is actually off-brand.
CBIZ ran into this directly: practitioners used Copilot to generate polished-looking presentations that weren’t on-brand, shared them with clients, and the issue only surfaced after the fact.
This is the DAM AI Governance Gap showing up in its most visible form: the tool worked exactly as designed, and the brand still ended up in a campaign looking like nobody owns it.
How do you fix this? Govern your AI the way you govern templates. Give AI access to approved brand kits and curated libraries, and build it into locked, protected structures rather than treating it as a free-standing tool anyone can point at a blank page.
The Payoff: Closing the Gap is a Growth Engine, Not Just Risk Reduction
It’s easy to frame all of this as risk reduction—protecting the brand, avoiding embarrassing mistakes—but that undersells it. Done well, closing the DAM AI Governance Gap doesn’t just prevent damage, it unlocks a level of scale no creative team could ever deliver by hand.
Here’s what that looks like in practice:
CBIZ manages its digital assets in the MediaValet digital asset management system, paired with MediaValet’s Templating, powered by Marq. Together, they let the team build governed, locked templates that pull directly from CRM and client data. So instead of filing a request and waiting on a designer, a tax professional can generate a client-ready presentation in a few clicks. The finished product is fully customized to that client, fully on-brand, and has no design bottleneck in the way.
At a company with 135,000 clients, there’s no design team on earth that could hand-build that volume of personalized material. Governance is precisely what makes it safe to automate in the first place: the template can’t be broken, so the output can’t go off-brand.
The adoption numbers back this up. Since launching the new template solution, roughly 90% of CBIZ’s users have adopted it within the first 90 days. That’s an adoption increase of an estimated 30–40% on the previous system.
The financial case holds too. By consolidating multiple point solutions into MediaValet and Marq, the rollout is effectively budget-neutral over the next few years.
There’s a quieter benefit as well: shifting repetitive, templated production work over to governed AI and automation frees designers to spend their time on the work that actually needs a designer, rather than squeezing the big creative project in at the last minute because routine requests ate the week.
Advice for Creative and Marketing Leaders Considering this Shift
Miller’s closing advice translates well beyond CBIZ’s specific situation:
- Get stakeholder buy-in before you go to leadership. Bringing an already-aligned group of stakeholders into a leadership conversation lands very differently than bringing a plan alone.
- Sell the pain-point relief, not the abstract brand value. “Better brand consistency” is a harder sell than “this cuts your turnaround time” or “you won’t need to wait on us for routine requests”.
- Look for connections outside your usual org chart. Shared data sources and shared tooling needs often show up in departments that never interact with creative day-to-day.
- Treat this as ongoing, not a one-time implementation. Technology changes, teams change, demand changes. AI governance has to keep adjusting with them, not get locked in place on launch day.
The organizations getting the most out of AI in creative operations aren’t necessarily the ones adopting it fastest. They’re the ones that fixed their governance and metadata foundations first so that AI has something solid to build on, instead of something fragile to expose.
Want the full conversation? Watch the complete DAM Week session with Ed Miller, Creative Director at CBIZ, and Steve Loewy, VP of Marketing at Marq, on demand.
